© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 15861
    Prev
    Next
    INDEX
    Explanations

    copyright and code headers

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron strongly activates on unusually long runs of whitespace (big gaps or indentation), i.e. large contiguous blocks of space characters.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    渽
    -2.41
    
    -2.41
    </b>
    -2.31
    ”،
    -2.28
    También
    -2.27
    一丝
    -2.25
    ༑
    -2.23
    ようになった
    -2.22
     ”
    -2.20
    0
    -2.17
    POSITIVE LOGITS
    \
    3.13
    的吧
    2.91
    大半
    2.67
     ,‬
    2.55
    ↵↵
    2.38
    !"
    2.34
    。
    2.30
    蔑
    2.28
    on
    2.28
    "
    2.27
    Activations Density 0.042%

    No Known Activations